Improvement of UAV Track Trajectory Algorithm Based on Ant Colony Algorithm

The track planning of the drone is to find the optimal flight route from the starting point to the target point to meet the maneuvering performance of the UAV and the requirements of the operational environment under certain constraints. In this paper, the corresponding threat constraint model is established for the constraints of the UAV flight path planning problem, including radar, missile, anti-aircraft and other threat models. The basic idea is to increase the target node pheromone guiding factor in the state transition of the basic ant colony algorithm, thereby reducing the blindness of the ant search and making the search toward the target node. At the same time, introducing the re-excitation learning mechanism, reasonable on the path The pheromone is renewed. At the end of each iteration, the individual behavior of ants is evaluated, and then the evaluation is fed back to the ant colony to promote the behavior of high quality ants and to punish the behavior of inferior ants, so as to realize the self-learning of ant colony on their own search behavior. After these improvements, the convergence speed and convergence ability of the algorithm are significantly improved.